
import pandas as pd
import numpy as np
from sklearn import metrics
import matplotlib.pyplot as plt
%matplotlib inline
import warnings
warnings.filterwarnings('ignore')
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.model_selection import GridSearchCV
from sklearn import metrics
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import AdaBoostClassifier, GradientBoostingClassifier, BaggingClassifier, RandomForestClassifier
from xgboost import XGBClassifier
import matplotlib as mpl
import IPython
IPython.display.set_matplotlib_formats('png2x')


# convenience function
def make_summary_cols(df, cat_threshold=20):
    types = df.dtypes
    types.name = 'dtypes'

    nuniques = data.nunique()
    nuniques.name = 'nuniques'

    nulls = df.isnull().sum()
    nulls.name = 'nulls'

    summary_cols = pd.merge(left=pd.merge(left=nuniques, right=types, left_index=True, right_index=True), right=nulls, left_index=True, right_index=True).sort_values(by='dtypes')
    summary_cols['isnumeric_column'] = summary_cols['dtypes'].apply(lambda x: False if x=='object' else True)

    summary_cols['probably_categorical'] = summary_cols['nuniques'].apply(lambda x: True if x <=cat_threshold else False)

    return summary_cols

def perc_on_bar(feature, data, figsize=[15,7],axes=None, hue=None):
    '''
    plot
    feature: categorical feature
    data: df to use
    figsize: set custom figsize
    axes: if setting up as multiplot
    hue: sns hue parameter
    '''
    #Creating a countplot for the feature
    sns.set(rc={'figure.figsize':tuple(figsize)})
    ax=sns.countplot(x=feature, data=data, ax=axes, hue=hue)
    
    total = len(feature) # length of the column
    for p in ax.patches:
        percentage = '{:.1f}%'.format(100 * p.get_height()/total) # percentage of each class of the category
        x = p.get_x() + p.get_width() / 2 - 0.25 # width of the plot
        y = p.get_y() + p.get_height()           # hieght of the plot
        ax.annotate(percentage, (x, y), size = 14) # annotate the percantage 
        
    #plt.show() # show the plot

def histogram_boxplot(feature, figsize=(15,10), bins = None):
    """ Boxplot and histogram combined
    feature: 1-d feature array
    figsize: size of fig (default (9,8))
    bins: number of bins (default None / auto)
    """
    f2, (ax_box2, ax_hist2) = plt.subplots(nrows = 2, # Number of rows of the subplot grid= 2
                                           sharex = True, # x-axis will be shared among all subplots
                                           gridspec_kw = {"height_ratios": (.25, .75)}, 
                                           figsize = figsize 
                                           ) # creating the 2 subplots
    sns.boxplot(feature, ax=ax_box2, showmeans=True, color='violet') # boxplot will be created and a star will indicate the mean value of the column
    sns.distplot(feature, kde=F, ax=ax_hist2, bins=bins,palette="winter") if bins else sns.distplot(feature, kde=False, ax=ax_hist2) # For histogram
    ax_hist2.axvline(np.mean(feature), color='green', linestyle='--') # Add mean to the histogram
    ax_hist2.axvline(np.median(feature), color='black', linestyle='-') # Add median to the histogram
    
    
def make_confusion_matrix(model,y_actual,labels=[1, 0], cmap='Blues'):
    '''
    model : classifier to predict values of X
    y_actual : ground truth  
    
    '''
    y_predict = model.predict(X_test)
    cm=metrics.confusion_matrix( y_actual, y_predict, labels=[0, 1])
    df_cm = pd.DataFrame(cm, index = [i for i in ["Actual - No","Actual - Yes"]],
                  columns = [i for i in ['Predicted - No','Predicted - Yes']])
    group_counts = ["{0:0.0f}".format(value) for value in
                cm.flatten()]
    group_percentages = ["{0:.2%}".format(value) for value in
                         cm.flatten()/np.sum(cm)]
    labels = [f"{v1}\n{v2}" for v1, v2 in
              zip(group_counts,group_percentages)]
    labels = np.asarray(labels).reshape(2,2)
    plt.figure(figsize = (10,7))
    sns.heatmap(df_cm, annot=labels,fmt='', cmap=cmap)
    plt.ylabel('True label')
    plt.xlabel('Predicted label')
    
    ##  Function to calculate accuracy score
    
def get_accuracy_score(model,flag=True):
    '''
    model : classifier to predict values of X

    '''
    c = [] # defining an empty list to store train and test results
    train_acc = model.score(X_train,y_train)
    test_acc = model.score(X_test,y_test)
    c.append(train_acc) # adding train accuracy to list
    c.append(test_acc) # adding test accuracy to list
    if flag == True: # If the flag is set to True then only the following print statements will be dispayed
        print("Accuracy on training set : ",model.score(X_train,y_train))
        print("Accuracy on test set : ",model.score(X_test,y_test))
    
    return c # returning the list with train and test scores


data = pd.read_excel('Tourism.xlsx', sheet_name='Tourism')
data_dict = pd.read_excel('Tourism.xlsx', sheet_name='Data Dict', skiprows=0,usecols='B:D') #using table within sheet
data_dict.columns = data_dict.loc[0] # set column names
data_dict = data_dict.loc[1:] #drop first row as contained col names
a = list(data_dict.columns[0:2])+['Description'] #correct spelling of third column
data_dict.columns = a #as above


data.head()


data_dict


data.shape

(4888, 20)


data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4888 entries, 0 to 4887
Data columns (total 20 columns):
 #   Column                   Non-Null Count  Dtype  
---  ------                   --------------  -----  
 0   CustomerID               4888 non-null   int64  
 1   ProdTaken                4888 non-null   int64  
 2   Age                      4662 non-null   float64
 3   PreferredLoginDevice     4863 non-null   object 
 4   CityTier                 4888 non-null   int64  
 5   DurationOfPitch          4637 non-null   float64
 6   Occupation               4888 non-null   object 
 7   Gender                   4888 non-null   object 
 8   NumberOfPersonVisited    4888 non-null   int64  
 9   NumberOfFollowups        4843 non-null   float64
 10  ProductPitched           4888 non-null   object 
 11  PreferredPropertyStar    4862 non-null   float64
 12  MaritalStatus            4888 non-null   object 
 13  NumberOfTrips            4748 non-null   float64
 14  Passport                 4888 non-null   int64  
 15  PitchSatisfactionScore   4888 non-null   int64  
 16  OwnCar                   4888 non-null   int64  
 17  NumberOfChildrenVisited  4822 non-null   float64
 18  Designation              4888 non-null   object 
 19  MonthlyIncome            4655 non-null   float64
dtypes: float64(7), int64(7), object(6)
memory usage: 763.9+ KB


data.describe()


data.drop('CustomerID', axis=1,inplace=True)
data


sns.pairplot(data, corner=True)

<seaborn.axisgrid.PairGrid at 0x7f881aaf8970>


sns.pairplot(data, hue='ProdTaken', corner=True)

<seaborn.axisgrid.PairGrid at 0x7f87fb692a30>


summary_cols = make_summary_cols(data)
display(summary_cols)
non_numeric_columns = list(summary_cols[summary_cols.isnumeric_column==False].index)
numeric_columns = list(summary_cols[summary_cols.isnumeric_column==True].index)


fig,axes = plt.subplots(figsize=(15,15))
sns.heatmap(data.loc[:,numeric_columns].corr(), annot=True, cmap='Blues')

<AxesSubplot:>


data.corr()


d = {}
for column in list(summary_cols[summary_cols.probably_categorical==True].index):
    #print(column)
    #data[column] = data[column].astype(np.cate .categorical)
    #print(data[column].unique())
    d[column]=data[column].unique()

for key in d:
    print(f'{key}: {d[key]}')

ProdTaken: [1 0]
OwnCar: [1 0]
CityTier: [3 1 2]
PitchSatisfactionScore: [2 3 5 4 1]
Passport: [1 0]
NumberOfPersonVisited: [3 2 1 4 5]
NumberOfChildrenVisited: [ 0.  2.  1. nan  3.]
NumberOfTrips: [ 1.  2.  7.  5.  6.  3.  4. 19. 21.  8. nan 20. 22.]
PreferredPropertyStar: [ 3.  4.  5. nan]
NumberOfFollowups: [ 3.  4.  2.  5. nan  1.  6.]
MaritalStatus: ['Single' 'Divorced' 'Married' 'Unmarried']
Gender: ['Female' 'Male' 'Fe Male']
Occupation: ['Salaried' 'Free Lancer' 'Small Business' 'Large Business']
PreferredLoginDevice: ['Self Enquiry' 'Company Invited' nan]
Designation: ['Manager' 'Executive' 'Senior Manager' 'AVP' 'VP']
ProductPitched: ['Super Deluxe' 'Multi' 'Standard' 'Deluxe' 'King']


# Fixing Gender
data.Gender = data.Gender.apply(lambda x: 'Female' if x=='Fe Male' else x)
data.Gender.unique()

array(['Female', 'Male'], dtype=object)


fig, axes = plt.subplots(1,3, figsize=(20,5))
for i,col in enumerate(['Age', 'MonthlyIncome', 'DurationOfPitch']):
    sns.distplot(data[col], ax=axes[i])
    #plt.show()


fig, axes = plt.subplots(1,3, figsize=(30,5))
for i,col in enumerate(['Age', 'MonthlyIncome', 'DurationOfPitch']):
    sns.distplot(data[data['ProdTaken']==0][col],color='orange',label=0, ax=axes[i])
    sns.distplot(data[data['ProdTaken']==1][col],color='indigo',label=1, ax=axes[i])
    plt.legend()
    #plt.show()

No handles with labels found to put in legend.
No handles with labels found to put in legend.


fig, axes = plt.subplots(1,3, figsize=(20,5))
for i,col in enumerate(['Age', 'MonthlyIncome', 'DurationOfPitch']):
    sns.kdeplot(data[col], ax=axes[i], hue=data['ProdTaken'])
    #plt.show()


data.loc[:,['Age', 'MonthlyIncome', 'DurationOfPitch', 'ProdTaken']].groupby('ProdTaken').mean()


data.loc[:,['Age', 'MonthlyIncome', 'DurationOfPitch', 'ProdTaken']].groupby('ProdTaken').median()


fig, axes = plt.subplots(4,4, figsize=(30,30))
i,j=0,0
for column in list(summary_cols[summary_cols.probably_categorical==True].index):
    #print(j,i)
    perc_on_bar(data[column], data, axes= axes[j,i], figsize=[30,30])
    i+=1
    if i==4:
        j+=1
        i=0


for column in list(summary_cols[summary_cols.probably_categorical==True].index):
    print(data[column].value_counts(normalize=True))
    print(' ')

0    0.811784
1    0.188216
Name: ProdTaken, dtype: float64
 
1    0.620295
0    0.379705
Name: OwnCar, dtype: float64
 
1    0.652619
3    0.306874
2    0.040507
Name: CityTier, dtype: float64
 
3    0.302373
5    0.198445
1    0.192717
4    0.186579
2    0.119885
Name: PitchSatisfactionScore, dtype: float64
 
0    0.709083
1    0.290917
Name: Passport, dtype: float64
 
3    0.491408
2    0.290098
4    0.209902
1    0.007979
5    0.000614
Name: NumberOfPersonVisited, dtype: float64
 
1.0    0.431356
2.0    0.276856
0.0    0.224388
3.0    0.067399
Name: NumberOfChildrenVisited, dtype: float64
 
2.0     0.308340
3.0     0.227254
1.0     0.130581
4.0     0.100674
5.0     0.096462
6.0     0.067818
7.0     0.045914
8.0     0.022115
20.0    0.000211
19.0    0.000211
22.0    0.000211
21.0    0.000211
Name: NumberOfTrips, dtype: float64
 
3.0    0.615590
5.0    0.196627
4.0    0.187783
Name: PreferredPropertyStar, dtype: float64
 
4.0    0.427008
3.0    0.302705
5.0    0.158579
2.0    0.047285
1.0    0.036341
6.0    0.028082
Name: NumberOfFollowups, dtype: float64
 
Married      0.478723
Divorced     0.194354
Single       0.187398
Unmarried    0.139525
Name: MaritalStatus, dtype: float64
 
Male      0.596563
Female    0.403437
Name: Gender, dtype: float64
 
Salaried          0.484452
Small Business    0.426350
Large Business    0.088789
Free Lancer       0.000409
Name: Occupation, dtype: float64
 
Self Enquiry       0.708205
Company Invited    0.291795
Name: PreferredLoginDevice, dtype: float64
 
Executive         0.376841
Manager           0.354337
Senior Manager    0.151800
AVP               0.069967
VP                0.047054
Name: Designation, dtype: float64
 
Multi           0.376841
Super Deluxe    0.354337
Standard        0.151800
Deluxe          0.069967
King            0.047054
Name: ProductPitched, dtype: float64


fig, axes = plt.subplots(4,4, figsize=(30,30))
i,j=0,0
for column in list(summary_cols[summary_cols.probably_categorical==True].index):
    #print(j,i)
    perc_on_bar(data[column], data, axes= axes[j,i], figsize=[30,30], hue='ProdTaken')
    i+=1
    if i==4:
        j+=1
        i=0


data.groupby('ProdTaken').mean()


for column in list(summary_cols[summary_cols.probably_categorical==True].index):
    display(pd.crosstab(data.ProdTaken,data[column], normalize='columns',margins=True))


data_prod = data[data.ProdTaken==1].groupby('ProductPitched') #We filter for those who actually took the product

x= data_prod.mean()
x


fig, axes = plt.subplots(1,2, figsize=(20,5))
x['Age'].plot(kind='bar', ax=axes[0], legend=True)
x['MonthlyIncome'].plot(kind='bar', ax=axes[1], legend=True)
del(x)


columns = ['PreferredLoginDevice', 'CityTier', 'Occupation', 'Gender', 'NumberOfPersonVisited', 'PreferredPropertyStar', 'MaritalStatus','NumberOfTrips','Passport','OwnCar','NumberOfChildrenVisited','Designation']

df_prod1 = data[data.ProdTaken==1]

for col in columns:
    display(pd.crosstab(df_prod1['ProductPitched'], df_prod1[col], normalize='index'))


data.groupby(['ProductPitched','Designation']).count()


# Copy to avoid corrupting original df in case need to undo changes
working_df = data.copy()


#no meaningful relation if we bin age
working_df['AgeBins'] = pd.cut(working_df.Age, bins=5, labels=range(0,5))
working_df.groupby('AgeBins').mean()['NumberOfChildrenVisited']

AgeBins
0    1.169978
1    1.197166
2    1.252882
3    1.245817
4    1.140162
Name: NumberOfChildrenVisited, dtype: float64


#Filling nans in Children - using mode as modal value is same across all marital status and mean is also ~1 across marital categories (no of children =1)
import math

display(working_df.groupby('MaritalStatus').mean()['NumberOfChildrenVisited'])

temp_df = working_df.copy()
# using 1000 as a placeholder value for NaN since crosstab ignores NaN
temp_df.NumberOfChildrenVisited = temp_df.NumberOfChildrenVisited.apply(lambda x: 1000 if math.isnan(x) else x)
display(pd.crosstab(temp_df.MaritalStatus, temp_df.NumberOfChildrenVisited, margins=True))
del(temp_df)



working_df.NumberOfChildrenVisited = working_df.NumberOfChildrenVisited.apply(lambda x: 1 if math.isnan(x) else x)

pd.crosstab(working_df.MaritalStatus, working_df.NumberOfChildrenVisited, margins=True)

MaritalStatus
Divorced     1.157556
Married      1.212462
Single       0.997768
Unmarried    1.391496
Name: NumberOfChildrenVisited, dtype: float64


#Fixing outliers and nan in NumberOfTrips
plt.figure(figsize=(20,5))
sns.countplot(working_df['NumberOfTrips'])
display(working_df.groupby('NumberOfTrips').mean())
display(working_df.groupby('NumberOfTrips').count())
#display(working_df[working_df.NumberOfTrips.isna()])


print('mean', working_df.NumberOfTrips.mean(), 'median', working_df.NumberOfTrips.median())

mean 3.236520640269587 median 3.0


temp_df = working_df[working_df.NumberOfTrips <=8]
print('mean where trips <=8:', temp_df.NumberOfTrips.mean(), 'median', temp_df.NumberOfTrips.median())

mean where trips <=8: 3.221964586846543 median 3.0


#No significant correlation with NumberOfTrips and any other var so we impute nan to median
#prefer median to mode as mode is 2 and NumberofTrips=3 is also significantly represented. Could also use mean but prefer using int values 
#mean=3.2 which is not strictly speaking possible although still possible to use for imputation to enable model fitting


from sklearn.impute import SimpleImputer
imp_mode = SimpleImputer(missing_values=np.nan, strategy='most_frequent')
imp_median = SimpleImputer(missing_values=np.nan, strategy='median')

#using only 

x=imp_median.fit_transform(working_df[['NumberOfTrips']])
s=pd.Series(x.flatten(),index=working_df.index)
s
working_df['NumberOfTrips'] = s

working_df


#working_df['NumberOfTripsOutlier'] = working_df['NumberOfTrips'].apply(lambda x: 1 if x > 8 else 0)
working_df[working_df['NumberOfTrips']>8].shape[0]

4


#Fixing nan in PreferredPropertyStar
plt.figure(figsize=(30,5))
sns.countplot(working_df['PreferredPropertyStar'])
display(working_df.groupby('PreferredPropertyStar').mean())
display(working_df.groupby('PreferredPropertyStar').count())
#display(working_df[working_df.PreferredPropertyStar.isna()])


print('mean', working_df.PreferredPropertyStar.mean(), 'median', working_df.PreferredPropertyStar.median())

mean 3.581036610448375 median 3.0


#No significant correlation with PreferredPropertyStar and any other var so we impute nan to mode

x=imp_mode.fit_transform(working_df[['PreferredPropertyStar']])
s=pd.Series(x.flatten(),index=working_df.index)
s
working_df['PreferredPropertyStar'] = s

working_df


#Fixing nan in NumberOfFollowups
plt.figure(figsize=(15,5))
sns.countplot(working_df['NumberOfFollowups'])
display(working_df.groupby('NumberOfFollowups').mean())
display(working_df.groupby('NumberOfFollowups').count())
#display(working_df[working_df.NumberOfFollowups.isna()])


#No significant correlation with NumberOfFollowups and any other var so we impute nan to mode

x=imp_mode.fit_transform(working_df[['NumberOfFollowups']])
s=pd.Series(x.flatten(),index=working_df.index)
s
working_df['NumberOfFollowups'] = s

working_df


#Fixing nan in PreferredLoginDevice
plt.figure(figsize=(15,5))
sns.countplot(working_df['PreferredLoginDevice'])
display(working_df.groupby('PreferredLoginDevice').mean())
display(working_df.groupby('PreferredLoginDevice').count())
#display(working_df[working_df.PreferredLoginDevice.isna()])


#No significant correlation with PreferredLoginDevice and any other var so we impute nan to mode

x=imp_mode.fit_transform(working_df[['PreferredLoginDevice']])
s=pd.Series(x.flatten(),index=working_df.index)
s
working_df['PreferredLoginDevice'] = s

#working_df


summary_cols2 = make_summary_cols(working_df)
display(summary_cols2)


histogram_boxplot(working_df.Age)


print('mean', working_df.Age.mean(), 'median', working_df.Age.median())

mean 37.62226512226512 median 36.0


working_df.groupby('AgeBins').mean()


working_df[working_df.Age.isna() & working_df.MonthlyIncome.isna()]
#while there is some correlation between Age and MonthlyIncome, 130 rows have both missing (out of total ~225-230 for each)
#we can see above that there is some correlation between AgeBins and NumberOfTrips but again it's very weak
#best to just use mean for age and monthlyincome


imp_mean = SimpleImputer(missing_values=np.nan, strategy='mean')

x=imp_mean.fit_transform(working_df[['Age']])
s=pd.Series(x.flatten(),index=working_df.index)
print(s)
working_df['Age'] = s

#x=imp_mean.fit_transform(working_df[['MonthlyIncome']])
#s=pd.Series(x.flatten(),index=working_df.index)
#s
#working_df['MonthlyIncome'] = s

#make_summary_cols(working_df)

0       41.000000
1       49.000000
2       37.000000
3       33.000000
4       37.622265
          ...    
4883    49.000000
4884    28.000000
4885    52.000000
4886    19.000000
4887    36.000000
Length: 4888, dtype: float64


histogram_boxplot(working_df.MonthlyIncome)


print('mean', working_df.MonthlyIncome.mean(), 'median', working_df.MonthlyIncome.median())

mean 23619.85349087003 median 22347.0


# Compute Z scores
m_inc = working_df.MonthlyIncome
print(m_inc)

# calculate mean
mean = np.mean(m_inc) # calculate standard deviation
sd = np.std(m_inc)# determine a threhold

z = (m_inc - mean)/sd
print(z)

working_df['MonthlyIncomeZ'] = z
working_df

0       20993.0
1       20130.0
2       17090.0
3       17909.0
4       18468.0
         ...   
4883    26576.0
4884    21212.0
4885    31820.0
4886    20289.0
4887    24041.0
Name: MonthlyIncome, Length: 4888, dtype: float64
0      -0.488252
1      -0.648657
2      -1.213700
3      -1.061473
4      -0.957572
          ...   
4883    0.549457
4884   -0.447546
4885    1.524157
4886   -0.619104
4887    0.078278
Name: MonthlyIncome, Length: 4888, dtype: float64


no_outlier_mean_m_inc = working_df[abs(working_df.MonthlyIncomeZ) < 3].mean()['MonthlyIncome']
no_outlier_median_m_inc = working_df[abs(working_df.MonthlyIncomeZ) < 3].median()['MonthlyIncome']

print('mean without outlier',no_outlier_mean_m_inc, 'median', no_outlier_median_m_inc)

mean without outlier 23597.304235648247 median 22347.0


print(f'No of rows with outlier {working_df[abs(working_df.MonthlyIncomeZ) >= 3].shape[0]}')

No of rows with outlier 4


x=imp_median.fit_transform(working_df[['MonthlyIncome']])
s=pd.Series(x.flatten(),index=working_df.index)
print(s)
working_df['MonthlyIncome'] = s

0       20993.0
1       20130.0
2       17090.0
3       17909.0
4       18468.0
         ...   
4883    26576.0
4884    21212.0
4885    31820.0
4886    20289.0
4887    24041.0
Length: 4888, dtype: float64


make_summary_cols(working_df)


histogram_boxplot(working_df.DurationOfPitch)


print('mean', working_df.DurationOfPitch.mean(), 'median', working_df.DurationOfPitch.median())

mean 15.490834591330602 median 13.0


sns.pairplot(working_df, y_vars=['DurationOfPitch'])

<seaborn.axisgrid.PairGrid at 0x7fd0fa4b55b0>


working_df['DurationBins']=pd.qcut(working_df.DurationOfPitch,5, labels=range(0,5))
working_df.groupby('DurationBins').mean()


# Compute Z scores
dur = working_df.DurationOfPitch
print(dur)

# calculate mean
mean = np.mean(dur) # calculate standard deviation
sd = np.std(dur)# determine a threhold

z = (dur - mean)/sd
print(z)

working_df['DurationOfPitchZ'] = z
working_df

0        6.0
1       14.0
2        8.0
3        9.0
4        8.0
        ... 
4883     9.0
4884    31.0
4885    17.0
4886    16.0
4887    14.0
Name: DurationOfPitch, Length: 4888, dtype: float64
0      -1.114115
1      -0.175007
2      -0.879338
3      -0.761949
4      -0.879338
          ...   
4883   -0.761949
4884    1.820597
4885    0.177159
4886    0.059770
4887   -0.175007
Name: DurationOfPitch, Length: 4888, dtype: float64


no_outlier_mean_dur = working_df[abs(working_df.DurationOfPitchZ) < 3].mean()['DurationOfPitch']
no_outlier_median_dur = working_df[abs(working_df.DurationOfPitchZ) < 3].median()['DurationOfPitch']

print('mean without outlier',no_outlier_mean_dur, 'median', no_outlier_median_dur)

mean without outlier 15.442934196332255 median 13.0


print(f'No of rows with outlier {working_df[abs(working_df.DurationOfPitchZ) >= 3].shape[0]}')

No of rows with outlier 2


x=imp_median.fit_transform(working_df[['DurationOfPitch']])
s=pd.Series(x.flatten(),index=working_df.index)
print(s)
working_df['DurationOfPitch'] = s

0        6.0
1       14.0
2        8.0
3        9.0
4        8.0
        ... 
4883     9.0
4884    31.0
4885    17.0
4886    16.0
4887    14.0
Length: 4888, dtype: float64


data['NumberOfPersonVisited'].value_counts()

3    2402
2    1418
4    1026
1      39
5       3
Name: NumberOfPersonVisited, dtype: int64


data['Occupation'].value_counts()

Salaried          2368
Small Business    2084
Large Business     434
Free Lancer          2
Name: Occupation, dtype: int64


working_df = working_df[working_df.Occupation!='Free Lancer']


working_df.columns

Index(['ProdTaken', 'Age', 'PreferredLoginDevice', 'CityTier',
       'DurationOfPitch', 'Occupation', 'Gender', 'NumberOfPersonVisited',
       'NumberOfFollowups', 'ProductPitched', 'PreferredPropertyStar',
       'MaritalStatus', 'NumberOfTrips', 'Passport', 'PitchSatisfactionScore',
       'OwnCar', 'NumberOfChildrenVisited', 'Designation', 'MonthlyIncome',
       'AgeBins', 'MonthlyIncomeZ', 'DurationBins', 'DurationOfPitchZ'],
      dtype='object')


working_df.drop(['AgeBins','DurationBins','MonthlyIncomeZ', 'DurationOfPitchZ'],axis=1,inplace=True)
working_df.columns

Index(['ProdTaken', 'Age', 'PreferredLoginDevice', 'CityTier',
       'DurationOfPitch', 'Occupation', 'Gender', 'NumberOfPersonVisited',
       'NumberOfFollowups', 'ProductPitched', 'PreferredPropertyStar',
       'MaritalStatus', 'NumberOfTrips', 'Passport', 'PitchSatisfactionScore',
       'OwnCar', 'NumberOfChildrenVisited', 'Designation', 'MonthlyIncome'],
      dtype='object')


working_df.drop('Designation', axis=1, inplace=True)


working_df


x = make_summary_cols(working_df)
display(x)
categorical_columns = list(x[x['dtypes']=='object'].index)
print(categorical_columns)
del(x)

['ProductPitched', 'Gender', 'MaritalStatus', 'Occupation', 'PreferredLoginDevice']


model_df = pd.get_dummies(working_df, columns=categorical_columns, drop_first=True)
model_df.columns

Index(['ProdTaken', 'Age', 'CityTier', 'DurationOfPitch',
       'NumberOfPersonVisited', 'NumberOfFollowups', 'PreferredPropertyStar',
       'NumberOfTrips', 'Passport', 'PitchSatisfactionScore', 'OwnCar',
       'NumberOfChildrenVisited', 'MonthlyIncome', 'ProductPitched_King',
       'ProductPitched_Multi', 'ProductPitched_Standard',
       'ProductPitched_Super Deluxe', 'Gender_Male', 'MaritalStatus_Married',
       'MaritalStatus_Single', 'MaritalStatus_Unmarried',
       'Occupation_Salaried', 'Occupation_Small Business',
       'PreferredLoginDevice_Self Enquiry'],
      dtype='object')


random_state = 314159


X = model_df.drop(['ProdTaken'],axis=1)
y = model_df['ProdTaken']
X_train, X_test, y_train, y_test =train_test_split(X, y, test_size=0.3, random_state=random_state,stratify=y)
print(X_train.shape, X_test.shape)

(3420, 23) (1466, 23)


print(f'Ratio of 1 in data: {y.value_counts()[1]/y.count()}')
print(f'Ratio of 1 in train data: {y_train.value_counts()[1]/y_train.count()}')
print(f'Ratio of 1 in test data: {y_test.value_counts()[1]/y_test.count()}')

Ratio of 1 in data: 0.18788374948833403
Ratio of 1 in train data: 0.18801169590643274
Ratio of 1 in test data: 0.18758526603001363


dtree= DecisionTreeClassifier(criterion='gini',random_state=random_state, class_weight='balanced')
dtree

DecisionTreeClassifier(class_weight='balanced', random_state=314159)


dtree.fit(X_train, y_train)

DecisionTreeClassifier(class_weight='balanced', random_state=314159)


make_confusion_matrix(dtree, y_test, cmap='coolwarm_r')


get_accuracy_score(dtree)

Accuracy on training set :  1.0
Accuracy on test set :  0.8922237380627558

[1.0, 0.8922237380627558]


y_predict = dtree.predict(X_test)
cm=metrics.confusion_matrix( y_test, y_predict, labels=[0, 1])
df_cm = pd.DataFrame(cm, index = [i for i in ["Actual - No","Actual - Yes"]], columns = [i for i in ['Predicted - No','Predicted - Yes']])
group_counts = ["{0:0.0f}".format(value) for value in cm.flatten()]
group_percentages = ["{0:.2%}".format(value) for value in cm.flatten()/np.sum(cm)]


def model_score_df(model, label='model', score_df=None):
    
        
    # Scores on Test Data
    y_predict = model.predict(X_test)
    
    
    TN, FP, FN, TP = metrics.confusion_matrix(y_test, y_predict).ravel()
    TNR = TN/(TN+FP)
    FPR = 1-TNR

    x={'model_accuracy':metrics.accuracy_score(y_test, y_predict),
       'model_recall_sensitivity_TPR':metrics.recall_score(y_test, y_predict),
       'model_precision':metrics.precision_score(y_test, y_predict),
       'Specificity_TNR':TNR,
       'model_f1':metrics.f1_score(y_test, y_predict),
       'FPR': FPR}
    
    df1 = pd.DataFrame(x, index=[label+'-test']).T
    
    #display(df1)
    # Scores on Train Data
    y_predict = model.predict(X_train)
    
    TN, FP, FN, TP = metrics.confusion_matrix(y_train, y_predict).ravel()
    TNR = TN/(TN+FP)
    FPR = 1-TNR

    x={'model_accuracy':metrics.accuracy_score(y_train, y_predict),
       'model_recall_sensitivity_TPR':metrics.recall_score(y_train, y_predict),
       'model_precision':metrics.precision_score(y_train, y_predict),
       'Specificity_TNR':TNR,
       'model_f1':metrics.f1_score(y_train, y_predict),
       'FPR': FPR}
    
    df2 = pd.DataFrame(x, index=[label+'-train']).T
    #display(df2)
    
    df_1_2 = pd.merge(df1, df2, left_index=True, right_index=True)
    
    if score_df is None:
        score_df = df_1_2
    else:
        score_df = pd.merge(score_df, df_1_2, left_index=True, right_index=True)
    
    
    return score_df


score_df = model_score_df(dtree, 'DT')
score_df


bagging = BaggingClassifier(random_state=random_state)
bagging.fit(X_train,y_train)
make_confusion_matrix(bagging, y_test, cmap='Reds')


get_accuracy_score(bagging)

Accuracy on training set :  0.9929824561403509
Accuracy on test set :  0.907912687585266

[0.9929824561403509, 0.907912687585266]


score_df = model_score_df(bagging, 'BG', score_df)
score_df


rf = RandomForestClassifier(random_state=random_state,class_weight='balanced')
rf.fit(X_train,y_train)

RandomForestClassifier(class_weight='balanced', random_state=314159)


make_confusion_matrix(rf, y_test)


get_accuracy_score(rf)

Accuracy on training set :  1.0
Accuracy on test set :  0.907912687585266

[1.0, 0.907912687585266]


score_df = model_score_df(rf, 'RF', score_df)
score_df


# Choose the type of classifier. 
dtree_estimator = DecisionTreeClassifier(random_state=random_state)

# Grid of parameters to choose from
parameters = {'max_depth': np.arange(2,30), 
              'min_samples_leaf': [1, 2, 5, 7, 10],
              'max_leaf_nodes' : [2, 3, 5, 10,15],
              'min_impurity_decrease': [0.00001,0.0001,0.001,0.01,0.1], 
              'class_weight':['balanced', None]
             }


# Run the grid search
grid_obj = GridSearchCV(dtree_estimator, parameters,n_jobs=-1)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
dtree_estimator = grid_obj.best_estimator_

# Fit the best algorithm to the data. 
dtree_estimator.fit(X_train, y_train)

DecisionTreeClassifier(max_depth=4, max_leaf_nodes=10,
                       min_impurity_decrease=1e-05, random_state=314159)


get_accuracy_score(dtree_estimator)

Accuracy on training set :  0.841812865497076
Accuracy on test set :  0.8376534788540245

[0.841812865497076, 0.8376534788540245]


make_confusion_matrix(dtree_estimator, y_test)


score_df = model_score_df(dtree_estimator, "DT-tuned", score_df)
score_df


# grid search for bagging classifier
cl1 = DecisionTreeClassifier(random_state=random_state)
param_grid = {'base_estimator':[cl1],
              'n_estimators':[5,7,15,51,101],
              'max_features': [0.7,0.8,0.9,1]
             }

grid = GridSearchCV(BaggingClassifier(random_state=1,bootstrap=True), param_grid=param_grid, cv = 5,n_jobs=-1)
grid.fit(X_train, y_train)

bagging_estimator  = grid.best_estimator_
bagging_estimator.fit(X_train,y_train)

BaggingClassifier(base_estimator=DecisionTreeClassifier(random_state=314159),
                  max_features=0.9, n_estimators=101, random_state=1)


make_confusion_matrix(bagging_estimator, y_test)


get_accuracy_score(bagging_estimator)

Accuracy on training set :  1.0
Accuracy on test set :  0.9229195088676672

[1.0, 0.9229195088676672]


score_df = model_score_df(bagging_estimator, 'BG-tuned', score_df)
score_df


# Choose the type of classifier. 
rf_estimator = RandomForestClassifier(random_state=random_state,oob_score=True,bootstrap=True)

# Grid of parameters to choose from
parameters = {  
                'max_features': ['sqrt','log2','auto'],
                'min_samples_leaf': np.arange(1,15,5),
                'min_samples_split': np.arange(2, 20, 5),
                'n_estimators': list(np.linspace(10, 50, 10, dtype = int)),
                'class_weight':['balanced',None]}


# Run the grid search
grid_obj = GridSearchCV(rf_estimator, parameters,cv=5,n_jobs=-1)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
rf_estimator = grid_obj.best_estimator_

# Fit the best algorithm to the data. 
rf_estimator.fit(X_train, y_train)

RandomForestClassifier(max_features='sqrt', n_estimators=45, oob_score=True,
                       random_state=314159)


make_confusion_matrix(rf_estimator, y_test)


get_accuracy_score(rf_estimator)

Accuracy on training set :  1.0
Accuracy on test set :  0.9133697135061392

[1.0, 0.9133697135061392]


score_df = model_score_df(rf_estimator, 'RF-tuned', score_df)
score_df


abc = AdaBoostClassifier(random_state=random_state)
abc.fit(X_train,y_train)

AdaBoostClassifier(random_state=314159)


make_confusion_matrix(abc, y_test)


get_accuracy_score(abc)

Accuracy on training set :  0.8538011695906432
Accuracy on test set :  0.8526603001364257

[0.8538011695906432, 0.8526603001364257]


score_df = model_score_df(abc, 'ABC', score_df)
score_df


gbc = GradientBoostingClassifier(random_state=random_state)
gbc.fit(X_train,y_train)

GradientBoostingClassifier(random_state=314159)


make_confusion_matrix(gbc, y_test)


get_accuracy_score(gbc)

Accuracy on training set :  0.8888888888888888
Accuracy on test set :  0.8724420190995907

[0.8888888888888888, 0.8724420190995907]


score_df = model_score_df(gbc, 'GBC', score_df)
score_df


xgb = XGBClassifier(random_state=random_state,eval_metric='logloss')
xgb.fit(X_train,y_train)

XGBClassifier(base_score=0.5, booster='gbtree', colsample_bylevel=1,
              colsample_bynode=1, colsample_bytree=1, eval_metric='logloss',
              gamma=0, gpu_id=-1, importance_type='gain',
              interaction_constraints='', learning_rate=0.300000012,
              max_delta_step=0, max_depth=6, min_child_weight=1, missing=nan,
              monotone_constraints='()', n_estimators=100, n_jobs=0,
              num_parallel_tree=1, random_state=314159, reg_alpha=0,
              reg_lambda=1, scale_pos_weight=1, subsample=1,
              tree_method='exact', validate_parameters=1, verbosity=None)


make_confusion_matrix(xgb, y_test)


get_accuracy_score(xgb)

Accuracy on training set :  0.9997076023391813
Accuracy on test set :  0.9242837653478854

[0.9997076023391813, 0.9242837653478854]


score_df = model_score_df(xgb, 'XGB', score_df)
score_df


# Choose the type of classifier. 
abc_tuned = AdaBoostClassifier(random_state=random_state)

# Grid of parameters to choose from
## add from article
parameters = {
    #Let's try different max_depth for base_estimator
    "base_estimator":[DecisionTreeClassifier(max_depth=1),DecisionTreeClassifier(max_depth=2),DecisionTreeClassifier(max_depth=3)],
    "n_estimators": np.arange(10,110,10),
    "learning_rate":np.arange(0.1,1,0.1)
}

# Run the grid search
grid_obj = GridSearchCV(abc_tuned, parameters,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
abc_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
abc_tuned.fit(X_train, y_train)


make_confusion_matrix(abc_tuned, y_test)


get_accuracy_score(abc_tuned)


score_df = model_score_df(abc_tuned, 'ABC-tuned', score_df)
score_df


# Choose the type of classifier. 
gbc_tuned = GradientBoostingClassifier(init=AdaBoostClassifier(random_state=random_state),random_state=random_state)

# Grid of parameters to choose from
## add from article
parameters = {
    "n_estimators": [100,150,200,250],
    "subsample":[0.8,0.9,1],
    "max_features":[0.7,0.8,0.9,1]
}

# Type of scoring used to compare parameter combinations


# Run the grid search
grid_obj = GridSearchCV(gbc_tuned, parameters,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
gbc_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
gbc_tuned.fit(X_train, y_train)


make_confusion_matrix(gbc_tuned, y_test)


get_accuracy_score(gbc_tuned)


score_df = model_score_df(gbc_tuned, 'GBC-tuned', score_df)
score_df


# Choose the type of classifier. 
xgb_tuned = XGBClassifier(random_state=random_state,eval_metric='logloss')

# Grid of parameters to choose from
## add from
parameters = {
    "n_estimators": np.arange(30,100,20),
    "subsample":[0.5,0.7,1],
    "learning_rate":[0.001,0.01,0.05,0.1],
    "colsample_bytree":[0.5,0.7,1],
    "colsample_bylevel":[0.5,0.7,1]
}

# Type of scoring used to compare parameter combinations

# Run the grid search
grid_obj = GridSearchCV(xgb_tuned, parameters,cv=5)
grid_obj = grid_obj.fit(X_train, y_train)

# Set the clf to the best combination of parameters
xgb_tuned = grid_obj.best_estimator_

# Fit the best algorithm to the data.
xgb_tuned.fit(X_train, y_train)


make_confusion_matrix(xgb_tuned, y_test)


get_accuracy_score(xgb_tuned)


score_df = model_score_df(xgb_tuned, 'XGB-tuned', score_df)
score_df


estimators=[('Decision Tree', dtree_tuned),('Random Forest', rf_tuned),
           ('Gradient Boosting', gbc_tuned)]
final_estimator=XGBClassifier(random_state=random_state,eval_metric='logloss')


stacking_estimator=StackingClassifier(estimators=estimators, final_estimator=final_estimator,cv=5)
stacking_estimator.fit(X_train,y_train)


make_confusion_matrix(stacking_estimator, y_test)


get_accuracy_score(stacking_estimator)


score_df = model_score_df(stacking_estimator, 'Stacking', score_df)
score_df


fig, axes = plt.subplots(2,2,figsize=(20,15))

feature_names = X_train.columns
importances = xgb.feature_importances_
indices = np.argsort(importances)

#plt.figure(figsize=(10,7))
plt.axes(axes[0,0])
plt.title('Feature Importances-XGB')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')
#plt.show()

feature_names = X_train.columns
importances = xgb_tuned.feature_importances_
indices = np.argsort(importances)

plt.axes(axes[0,1])
plt.title('Feature Importances-XGB tuned')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')

feature_names = X_train.columns
importances = gbc.feature_importances_
indices = np.argsort(importances)

plt.axes(axes[1,0])
plt.title('Feature Importances-GBC')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')

feature_names = X_train.columns
importances = gbc_tuned.feature_importances_
indices = np.argsort(importances)

plt.axes(axes[1,1])
plt.title('Feature Importances-GBC tuned')
plt.barh(range(len(indices)), importances[indices], color='violet', align='center')
plt.yticks(range(len(indices)), [feature_names[i] for i in indices])
plt.xlabel('Relative Importance')


score_df.T


a = list(np.arange(1,score_df.shape[1],2))

x=score_df.T.iloc[a]
x = x[x.model_accuracy<=0.98]
display(x)

no_overf_models = list(x.index)
print(no_overf_models)


b=[]
for a in no_overf_models:
    b.append(a.replace('train', 'test'))
    
score_df.T.loc[b]

	CustomerID	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
count	4888.000000	4888.000000	4662.000000	4888.000000	4637.000000	4888.000000	4843.000000	4862.000000	4748.000000	4888.000000	4888.000000	4888.000000	4822.000000	4655.000000
mean	202443.500000	0.188216	37.622265	1.654255	15.490835	2.905074	3.708445	3.581037	3.236521	0.290917	3.078151	0.620295	1.187267	23619.853491
std	1411.188388	0.390925	9.316387	0.916583	8.519643	0.724891	1.002509	0.798009	1.849019	0.454232	1.365792	0.485363	0.857861	5380.698361
min	200000.000000	0.000000	18.000000	1.000000	5.000000	1.000000	1.000000	3.000000	1.000000	0.000000	1.000000	0.000000	0.000000	1000.000000
25%	201221.750000	0.000000	31.000000	1.000000	9.000000	2.000000	3.000000	3.000000	2.000000	0.000000	2.000000	0.000000	1.000000	20346.000000
50%	202443.500000	0.000000	36.000000	1.000000	13.000000	3.000000	4.000000	3.000000	3.000000	0.000000	3.000000	1.000000	1.000000	22347.000000
75%	203665.250000	0.000000	44.000000	3.000000	20.000000	3.000000	4.000000	4.000000	4.000000	1.000000	4.000000	1.000000	2.000000	25571.000000
max	204887.000000	1.000000	61.000000	3.000000	127.000000	5.000000	6.000000	5.000000	22.000000	1.000000	5.000000	1.000000	3.000000	98678.000000

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
ProdTaken	1.000000	-0.147254	0.086852	0.078257	0.009627	0.112171	0.099577	0.018898	0.260844	0.051394	-0.011508	0.007421	-0.130585
Age	-0.147254	1.000000	-0.015625	-0.012063	0.011621	-0.002577	-0.010474	0.184905	0.033399	0.018510	0.048654	0.007370	0.464869
CityTier	0.086852	-0.015625	1.000000	0.022703	-0.001671	0.023652	-0.009164	-0.029709	0.001793	-0.042160	0.003817	0.000672	0.051817
DurationOfPitch	0.078257	-0.012063	0.022703	1.000000	0.065141	0.009434	-0.006637	0.009715	0.033034	-0.002880	-0.001626	0.031408	-0.006252
NumberOfPersonVisited	0.009627	0.011621	-0.001671	0.065141	1.000000	0.328569	0.033867	0.195223	0.011177	-0.019581	0.010362	0.610621	0.195134
NumberOfFollowups	0.112171	-0.002577	0.023652	0.009434	0.328569	1.000000	-0.024176	0.139517	0.004970	0.004054	0.012112	0.286425	0.176503
PreferredPropertyStar	0.099577	-0.010474	-0.009164	-0.006637	0.033867	-0.024176	1.000000	0.012115	0.001040	-0.022701	0.015742	0.035798	0.014289
NumberOfTrips	0.018898	0.184905	-0.029709	0.009715	0.195223	0.139517	0.012115	1.000000	0.012949	-0.004378	-0.011825	0.168795	0.139105
Passport	0.260844	0.033399	0.001793	0.033034	0.011177	0.004970	0.001040	0.012949	1.000000	0.002926	-0.022330	0.020264	0.002545
PitchSatisfactionScore	0.051394	0.018510	-0.042160	-0.002880	-0.019581	0.004054	-0.022701	-0.004378	0.002926	1.000000	0.068850	0.000878	0.030421
OwnCar	-0.011508	0.048654	0.003817	-0.001626	0.010362	0.012112	0.015742	-0.011825	-0.022330	0.068850	1.000000	0.026572	0.080262
NumberOfChildrenVisited	0.007421	0.007370	0.000672	0.031408	0.610621	0.286425	0.035798	0.168795	0.020264	0.000878	0.026572	1.000000	0.201643
MonthlyIncome	-0.130585	0.464869	0.051817	-0.006252	0.195134	0.176503	0.014289	0.139105	0.002545	0.030421	0.080262	0.201643	1.000000

	Age	MonthlyIncome	DurationOfPitch
ProdTaken
0	38.282092	23960.962835	15.169325
1	34.770548	22172.827703	16.873143

	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
ProdTaken
0	38.282092	1.615927	15.169325	2.901714	3.654286	3.542806	3.219531	0.233871	3.044355	0.622984	1.184190	23960.962835
1	34.770548	1.819565	16.873143	2.919565	3.941886	3.746171	3.308370	0.536957	3.223913	0.608696	1.200438	22172.827703

OwnCar	0	1	All
ProdTaken
0	0.806034	0.815303	0.811784
1	0.193966	0.184697	0.188216

Visualisations - Continuous/ Discrete Non Categorical

Visualisations - Continuous/ Discrete non Categorical, Bivariate

Visualisations - Categorical

Visualisation - Categorical, Bivariate

Customer Profile vs. Product Pitched

Customer profiles by pitched product

Filling NaNs/ Other Analysis

Children NaNs

Number of Trips NaNs & outliers

Pref Prop Star NaNs

NaN in NumberOfFollowups

NaN in PreferredLoginDevice

NaN in Age

NaN in Monthly Income

NaN in Duration of Pitch

Outliers in Categorical Columns

Cleanup columns

One Hot For Categorical non numeric columns

Model Building

Decision Tree, Bagging, Random Forest

Tuning Dtree, RF and Bagging

Adaboost¶

Gradient Boosting¶

XGBoost¶

Tuning Adaboost, Gradient and XG

Tuning Gradient Boosting

Tuning XGBoost

Stacking Classifier

Models with no overfitting

Business implications

Best Model

Best Features

Metrics

Output if we dont drop rows with occupation=Free Lancer

Output with Data including Occupation=Free Lancer

	CustomerID	ProdTaken	Age	PreferredLoginDevice	CityTier	DurationOfPitch	Occupation	Gender	NumberOfPersonVisited	NumberOfFollowups	ProductPitched	PreferredPropertyStar	MaritalStatus	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	Designation	MonthlyIncome
0	200000	1	41.0	Self Enquiry	3	6.0	Salaried	Female	3	3.0	Super Deluxe	3.0	Single	1.0	1	2	1	0.0	Manager	20993.0
1	200001	0	49.0	Company Invited	1	14.0	Salaried	Male	3	4.0	Super Deluxe	4.0	Divorced	2.0	0	3	1	2.0	Manager	20130.0
2	200002	1	37.0	Self Enquiry	1	8.0	Free Lancer	Male	3	4.0	Multi	3.0	Single	7.0	1	3	0	0.0	Executive	17090.0
3	200003	0	33.0	Company Invited	1	9.0	Salaried	Female	2	3.0	Multi	3.0	Divorced	2.0	1	5	1	1.0	Executive	17909.0
4	200004	0	NaN	Self Enquiry	1	8.0	Small Business	Male	2	3.0	Multi	4.0	Divorced	1.0	0	5	1	0.0	Executive	18468.0

	Data	Variable	Description
1	Tourism	CustomerID	Unique customer ID
2	Tourism	ProdTaken	Product taken flag
3	Tourism	Age	Age of customer
4	Tourism	PreferredLoginDevice	Preferred login device of customer in last month
5	Tourism	CityTier	City tier
6	Tourism	DurationOfPitch	Duration of pitch by a sales man to customer
7	Tourism	Occupation	Occupation of customer
8	Tourism	Gender	Gender of customer
9	Tourism	NumberOfPersonVisited	Total number of person came with customer
10	Tourism	NumberOfFollowups	Total number of follow up has been done by sal...
11	Tourism	ProductPitched	Product pitched by sales person
12	Tourism	PreferredPropertyStar	Preferred hotel property rating by customer
13	Tourism	MaritalStatus	Marital status of customer
14	Tourism	NumberOfTrips	Average number of trip in a year by customer
15	Tourism	Passport	Customer passport flag
16	Tourism	PitchSatisfactionScore	Sales pitch satisfactory score
17	Tourism	OwnCar	Customers owns a car flag
18	Tourism	NumberOfChildrenVisited	Total number of children visit with customer
19	Tourism	Designation	Designation of customer in current organization
20	Tourism	MonthlyIncome	Gross monthly income of customer

	nuniques	dtypes	nulls	isnumeric_column	probably_categorical
ProdTaken	2	int64	0	True	True
OwnCar	2	int64	0	True	True
CityTier	3	int64	0	True	True
PitchSatisfactionScore	5	int64	0	True	True
Passport	2	int64	0	True	True
NumberOfPersonVisited	5	int64	0	True	True
NumberOfChildrenVisited	4	float64	66	True	True
NumberOfTrips	12	float64	140	True	True
PreferredPropertyStar	3	float64	26	True	True
MonthlyIncome	2475	float64	233	True	False
NumberOfFollowups	6	float64	45	True	True
DurationOfPitch	34	float64	251	True	False
Age	44	float64	226	True	False
MaritalStatus	4	object	0	False	True
Gender	3	object	0	False	True
Occupation	4	object	0	False	True
PreferredLoginDevice	2	object	25	False	True
Designation	5	object	0	False	True
ProductPitched	5	object	0	False	True

CityTier	1	2	3	All
ProdTaken
0	0.836991	0.767677	0.764	0.811784
1	0.163009	0.232323	0.236	0.188216

PitchSatisfactionScore	1	2	3	4	5	All
ProdTaken
0	0.847134	0.849829	0.786198	0.822368	0.783505	0.811784
1	0.152866	0.150171	0.213802	0.177632	0.216495	0.188216

NumberOfPersonVisited	1	2	3	4	5	All
ProdTaken
0	1.0	0.811707	0.808493	0.811891	1.0	0.811784
1	0.0	0.188293	0.191507	0.188109	0.0	0.188216

NumberOfChildrenVisited	0.0	1.0	2.0	3.0	All
ProdTaken
0	0.813309	0.811538	0.810487	0.796923	0.810659
1	0.186691	0.188462	0.189513	0.203077	0.189341

NumberOfTrips	1.0	2.0	3.0	4.0	5.0	6.0	7.0	8.0	19.0	20.0	21.0	22.0	All
ProdTaken
0	0.819355	0.795765	0.798888	0.872385	0.864629	0.801242	0.715596	0.72381	0.0	0.0	1.0	1.0	0.808762
1	0.180645	0.204235	0.201112	0.127615	0.135371	0.198758	0.284404	0.27619	1.0	1.0	0.0	0.0	0.191238

PreferredPropertyStar	3.0	4.0	5.0	All
ProdTaken
0	0.838958	0.800657	0.738494	0.812012
1	0.161042	0.199343	0.261506	0.187988

NumberOfFollowups	1.0	2.0	3.0	4.0	5.0	6.0	All
ProdTaken
0	0.886364	0.895197	0.833561	0.816731	0.751302	0.602941	0.811687
1	0.113636	0.104803	0.166439	0.183269	0.248698	0.397059	0.188313

Designation	AVP	Executive	Manager	Senior Manager	VP	All
ProdTaken
0	0.94152	0.700326	0.882217	0.832884	0.913043	0.811784
1	0.05848	0.299674	0.117783	0.167116	0.086957	0.188216

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
ProductPitched
Deluxe	1.0	43.500000	2.600000	18.500000	2.700000	3.100000	3.600000	3.263158	0.600000	3.800000	1.000000	1.200000	29823.800000
King	1.0	48.900000	1.800000	10.500000	2.900000	4.300000	3.750000	3.411765	0.600000	3.300000	0.900000	1.437500	34672.100000
Multi	1.0	31.289320	1.510870	15.791353	2.905797	3.950730	3.771739	3.226691	0.583333	3.213768	0.572464	1.221416	20161.529301
Standard	1.0	41.008130	2.096774	19.065041	2.967742	3.935484	3.731707	3.016260	0.387097	3.467742	0.661290	1.121951	26035.419355
Super Deluxe	1.0	37.641414	2.411765	19.100000	2.950980	3.970000	3.699507	3.702970	0.490196	3.039216	0.607843	1.172414	23106.215385

CityTier	1	2	3
ProductPitched
Deluxe	0.200000	0.000000	0.800000
King	0.600000	0.000000	0.400000
Multi	0.710145	0.068841	0.221014
Standard	0.419355	0.064516	0.516129
Super Deluxe	0.294118	0.000000	0.705882

Gender	Female	Male
ProductPitched
Deluxe	0.200000	0.800000
King	0.600000	0.400000
Multi	0.376812	0.623188
Standard	0.387097	0.612903
Super Deluxe	0.343137	0.656863

MaritalStatus	Divorced	Married	Single	Unmarried	All
ProdTaken
0	0.869474	0.860684	0.668122	0.756598	0.811784
1	0.130526	0.139316	0.331878	0.243402	0.188216

Occupation	Free Lancer	Large Business	Salaried	Small Business	All
ProdTaken
0	0.0	0.723502	0.825169	0.815739	0.811784
1	1.0	0.276498	0.174831	0.184261	0.188216

NumberOfTrips	1.0	2.0	3.0	4.0	5.0	6.0	7.0	8.0	19.0	20.0
ProductPitched
Deluxe	0.315789	0.315789	0.000000	0.000000	0.105263	0.105263	0.105263	0.052632	0.000000	0.000000
King	0.117647	0.235294	0.352941	0.058824	0.000000	0.117647	0.117647	0.000000	0.000000	0.000000
Multi	0.102377	0.361974	0.283364	0.049360	0.047532	0.053016	0.063985	0.034735	0.001828	0.001828
Standard	0.162602	0.341463	0.235772	0.081301	0.040650	0.056911	0.056911	0.024390	0.000000	0.000000
Super Deluxe	0.138614	0.242574	0.133663	0.113861	0.143564	0.118812	0.079208	0.029703	0.000000	0.000000

Product Type	Mean Age	Mean Monthly Income	Preferred Login	City Tier	Occupation	Gender	No of Person Visited	Pref Prop Star	Marital Status	No of Trips	Passport	Own Car	No of Children	Desig
Multi	31.3	20162	Self Enquiry (65%)	1-71%	Salaried-47%, Small Bus 37%	Male 62%	3-50%, 2-30%	3-51%, 5-28%	Single-42%, Married-34%	2-36%, 3-28%	1-58%	1-57%	1-42%, 2-29%	Executive-100%
Deluxe	43.5	29824	Company Invited (80%)	3-80%	Salaried-80%	Male 80%	3-50%, 2-40%	3-60%, 4,5-20%	Single-50%, Married-40%	1,2-32%	1-60%	1-100%	1-45%, 2-30%	AVP-100%
King	48.9	34672	Self Enquiry (100%)	1-60%	Small Bus 60%	Female 60%	3-50%, 2-30%	4-50%,3-37.5%	Single-40%, Married, Divorced-30%	3-35%, 2-24%	1-60%	1-90%	1-44%, 2-31%	VP-100%
Standard	41	26035	Self Enquiry (74%)	3-51%	Small Bus 47%, Salaried 44%	Male 61%	3-50%, 2-27%	3-55%, 5-28%	Married-45%, Unmarried-25%	2-34%, 3-24%	0-61%	1-66%	1-42%, 0-27%	Senior Manager-100%
Super Deluxe	37.6	23106	Self Enquiry (67%)	3-71%	Small Bus 53%, Salaried 39%	Male 66%	3-50%, 2-27%	3-56%, 5-26%	Married-33%, Unmarried-29%	2-24%	0-51%	1-61%	1-44%, 2-28%	Manager-100%

NumberOfChildrenVisited	0.0	1.0	2.0	3.0	1000.0	All
MaritalStatus
Divorced	226	400	241	66	17	950
Married	498	987	663	163	29	2340
Single	276	383	200	37	20	916
Unmarried	82	310	231	59	0	682
All	1082	2080	1335	325	66	4888

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
NumberOfTrips
1.0	0.180645	37.947368	1.693548	14.370968	2.401613	3.181818	3.609047	0.267742	3.166129	0.625806	0.685484	22048.389680
2.0	0.204235	34.800727	1.680328	15.507236	2.814208	3.614698	3.570548	0.284153	3.066940	0.635246	1.081284	22365.698901
3.0	0.201112	34.782732	1.653383	16.067943	3.177943	4.003749	3.574488	0.297498	3.009268	0.621872	1.452271	23432.665720
4.0	0.127615	42.002151	1.629707	15.909292	2.830544	3.671548	3.568134	0.320084	3.098326	0.564854	1.112971	24576.316703
5.0	0.135371	41.579065	1.615721	16.041667	2.903930	3.665929	3.516630	0.325328	3.161572	0.593886	1.183406	24086.855856
6.0	0.198758	40.954839	1.673913	15.987013	3.071429	3.822981	3.615142	0.288820	2.990683	0.627329	1.369565	24205.265574
7.0	0.284404	38.918660	1.642202	14.786730	2.935780	3.789720	3.699074	0.256881	3.022936	0.619266	1.201835	23562.361502
8.0	0.276190	39.057143	1.533333	13.904762	3.495238	4.323810	3.653846	0.266667	3.257143	0.609524	1.771429	24621.704762
19.0	1.000000	30.000000	1.000000	10.000000	2.000000	3.000000	3.000000	1.000000	4.000000	1.000000	1.000000	17285.000000
20.0	1.000000	31.000000	1.000000	11.000000	3.000000	4.000000	3.000000	1.000000	4.000000	1.000000	2.000000	20963.000000
21.0	0.000000	39.000000	1.000000	15.000000	3.000000	3.000000	4.000000	0.000000	2.000000	1.000000	0.000000	21782.000000
22.0	0.000000	40.000000	1.000000	16.000000	4.000000	4.000000	4.000000	0.000000	2.000000	1.000000	1.000000	25460.000000

	ProdTaken	Age	PreferredLoginDevice	CityTier	DurationOfPitch	Occupation	Gender	NumberOfPersonVisited	NumberOfFollowups	ProductPitched	PreferredPropertyStar	MaritalStatus	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	Designation	MonthlyIncome	AgeBins
NumberOfTrips
1.0	620	551	614	620	558	620	620	620	605	620	619	620	620	620	620	620	620	562	551
2.0	1464	1375	1456	1464	1382	1464	1464	1464	1456	1464	1460	1464	1464	1464	1464	1464	1464	1365	1375
3.0	1079	1054	1076	1079	1045	1079	1079	1079	1067	1079	1074	1079	1079	1079	1079	1079	1079	1056	1054
4.0	478	465	475	478	452	478	478	478	478	478	477	478	478	478	478	478	478	461	465
5.0	458	449	456	458	432	458	458	458	452	458	451	458	458	458	458	458	458	444	449
6.0	322	310	321	322	308	322	322	322	322	322	317	322	322	322	322	322	322	305	310
7.0	218	209	216	218	211	218	218	218	214	218	216	218	218	218	218	218	218	213	209
8.0	105	105	105	105	105	105	105	105	105	105	104	105	105	105	105	105	105	105	105
19.0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1
20.0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1
21.0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1
22.0	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
PreferredPropertyStar
3.0	0.161042	37.670403	1.655864	15.527709	2.884063	3.718908	3.203809	0.295356	3.085533	0.612429	1.157033	23521.668765
4.0	0.199343	37.174857	1.685652	15.536782	2.914567	3.708749	3.258488	0.260679	3.141292	0.628697	1.212486	23227.312285
5.0	0.261506	37.533113	1.623431	15.363436	2.945607	3.652587	3.251046	0.306485	2.981172	0.629707	1.225941	23819.335929

	ProdTaken	Age	PreferredLoginDevice	CityTier	DurationOfPitch	Occupation	Gender	NumberOfPersonVisited	NumberOfFollowups	ProductPitched	MaritalStatus	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	Designation	MonthlyIncome	AgeBins
PreferredPropertyStar
3.0	2993	2855	2974	2993	2833	2993	2993	2993	2967	2993	2993	2993	2993	2993	2993	2993	2993	2859	2855
4.0	913	875	912	913	870	913	913	913	903	913	913	913	913	913	913	913	913	871	875
5.0	956	906	951	956	908	956	956	956	947	956	956	956	956	956	956	956	956	899	906

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
NumberOfFollowups
1.0	0.113636	37.381250	1.568182	15.248447	2.437500	3.630682	2.744318	0.301136	2.875000	0.613636	0.704545	21239.670732
2.0	0.104803	37.892857	1.558952	16.180180	3.205240	3.655022	3.475983	0.301310	2.930131	0.606987	1.528384	24064.915929
3.0	0.166439	37.490950	1.654843	15.220957	2.449523	3.577080	2.789905	0.286494	3.120055	0.615280	0.718963	22019.930904
4.0	0.183269	37.839126	1.661025	15.624378	3.096228	3.575919	3.403772	0.289168	3.107350	0.620406	1.368956	24342.400099
5.0	0.248698	37.371202	1.652344	15.358289	3.217448	3.572917	3.562500	0.299479	3.029948	0.631510	1.515625	24838.099338
6.0	0.397059	37.250000	1.735294	16.552239	3.389706	3.433824	3.808824	0.316176	2.904412	0.632353	1.691176	24998.036765

	ProdTaken	Age	PreferredLoginDevice	CityTier	DurationOfPitch	Occupation	Gender	NumberOfPersonVisited	ProductPitched	PreferredPropertyStar	MaritalStatus	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	Designation	MonthlyIncome	AgeBins
NumberOfFollowups
1.0	176	160	172	176	161	176	176	176	176	176	176	176	176	176	176	176	176	164	160
2.0	229	224	229	229	222	229	229	229	229	229	229	229	229	229	229	229	229	226	224
3.0	1466	1326	1456	1466	1317	1466	1466	1466	1466	1466	1466	1466	1466	1466	1466	1466	1466	1317	1326
4.0	2068	2014	2061	2068	2010	2068	2068	2068	2068	2068	2068	2068	2068	2068	2068	2068	2068	2012	2014
5.0	768	757	764	768	748	768	768	768	768	768	768	768	768	768	768	768	768	755	757
6.0	136	136	136	136	134	136	136	136	136	136	136	136	136	136	136	136	136	136	136

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
PreferredLoginDevice
Company Invited	0.218464	37.286344	1.668076	15.894303	2.909091	3.736434	3.539112	3.198732	0.290345	3.116984	0.618041	1.194503	23390.292068
Self Enquiry	0.176249	37.819542	1.648664	15.327884	2.907666	3.703833	3.594948	3.244774	0.293264	3.064170	0.621661	1.184959	23713.525106

	ProdTaken	Age	CityTier	DurationOfPitch	Occupation	Gender	NumberOfPersonVisited	NumberOfFollowups	ProductPitched	PreferredPropertyStar	MaritalStatus	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	Designation	MonthlyIncome	AgeBins
PreferredLoginDevice
Company Invited	1419	1362	1419	1334	1419	1419	1419	1419	1419	1419	1419	1419	1419	1419	1419	1419	1419	1349	1362
Self Enquiry	3444	3275	3444	3303	3444	3444	3444	3444	3444	3444	3444	3444	3444	3444	3444	3444	3444	3306	3275

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome
AgeBins
0	0.397351	23.181015	1.728477	16.021792	2.867550	3.644592	3.578366	2.571744	0.320088	3.178808	0.567329	1.169978	19708.016018
1	0.211889	31.357858	1.687463	15.489281	2.932902	3.770453	3.589170	3.007652	0.257799	3.035315	0.604473	1.196586	21922.054579
2	0.119392	39.123954	1.628897	15.424581	2.953612	3.732319	3.577186	3.519392	0.294297	2.996958	0.630418	1.250190	24266.525869
3	0.150498	47.603234	1.671642	15.442581	2.962687	3.753731	3.555970	3.597015	0.318408	3.174129	0.645522	1.237562	26804.689394
4	0.148338	56.000000	1.639386	15.289062	2.879795	3.682864	3.557545	3.677749	0.327366	3.161125	0.659847	1.132992	27862.619171

	ProdTaken	Age	PreferredLoginDevice	CityTier	DurationOfPitch	Occupation	Gender	NumberOfPersonVisited	NumberOfFollowups	ProductPitched	...	MaritalStatus	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	Designation	MonthlyIncome	AgeBins	MonthlyIncomeZ
0	1	41.000000	Self Enquiry	3	6.0	Salaried	Female	3	3.0	Super Deluxe	...	Single	1.0	1	2	1	0.0	Manager	20993.0	2	-0.488252
1	0	49.000000	Company Invited	1	14.0	Salaried	Male	3	4.0	Super Deluxe	...	Divorced	2.0	0	3	1	2.0	Manager	20130.0	3	-0.648657
2	1	37.000000	Self Enquiry	1	8.0	Free Lancer	Male	3	4.0	Multi	...	Single	7.0	1	3	0	0.0	Executive	17090.0	2	-1.213700
3	0	33.000000	Company Invited	1	9.0	Salaried	Female	2	3.0	Multi	...	Divorced	2.0	1	5	1	1.0	Executive	17909.0	1	-1.061473
4	0	37.622265	Self Enquiry	1	8.0	Small Business	Male	2	3.0	Multi	...	Divorced	1.0	0	5	1	0.0	Executive	18468.0	NaN	-0.957572
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
4883	1	49.000000	Self Enquiry	3	9.0	Small Business	Male	3	5.0	Super Deluxe	...	Unmarried	2.0	1	1	1	1.0	Manager	26576.0	3	0.549457
4884	1	28.000000	Company Invited	1	31.0	Salaried	Male	4	5.0	Multi	...	Single	3.0	1	3	1	2.0	Executive	21212.0	1	-0.447546
4885	1	52.000000	Self Enquiry	3	17.0	Salaried	Female	4	4.0	Standard	...	Married	7.0	0	1	1	3.0	Senior Manager	31820.0	3	1.524157
4886	1	19.000000	Self Enquiry	3	16.0	Small Business	Male	3	4.0	Multi	...	Single	3.0	0	5	0	2.0	Executive	20289.0	0	-0.619104
4887	1	36.000000	Self Enquiry	1	14.0	Salaried	Male	4	4.0	Multi	...	Unmarried	3.0	1	3	1	2.0	Executive	24041.0	2	0.078278

	ProdTaken	Age	CityTier	DurationOfPitch	NumberOfPersonVisited	NumberOfFollowups	PreferredPropertyStar	NumberOfTrips	Passport	PitchSatisfactionScore	OwnCar	NumberOfChildrenVisited	MonthlyIncome	MonthlyIncomeZ
DurationBins
0	0.148785	38.499306	1.580972	7.014170	2.787449	3.629555	3.606275	3.053644	0.257085	3.028340	0.610324	1.079960	23389.126518	-0.034499
1	0.172747	37.990369	1.700644	9.701717	2.989270	3.863734	3.530043	3.451717	0.286481	3.136266	0.623391	1.268240	24416.068670	0.160343
2	0.168085	37.593902	1.615957	13.634043	2.893617	3.684043	3.590426	3.330851	0.284043	3.032979	0.629787	1.195745	23516.287234	-0.014284
3	0.225113	36.886905	1.704751	18.494344	3.020362	3.797511	3.624434	3.209276	0.324661	3.169683	0.641403	1.300905	23661.739819	0.014313
4	0.235162	37.862264	1.634938	29.892497	2.977604	3.709966	3.544233	3.236282	0.318029	3.035834	0.589026	1.207167	23595.592385	0.001352

	DT-test	DT-train
model_accuracy	0.892224	1.0
model_recall_sensitivity_TPR	0.709091	1.0
model_precision	0.714286	1.0
Specificity_TNR	0.934509	1.0
model_f1	0.711679	1.0
FPR	0.065491	0.0

	DT-test	DT-train	BG-test	BG-train
model_accuracy	0.892224	1.0	0.907913	0.992982
model_recall_sensitivity_TPR	0.709091	1.0	0.563636	0.964230
model_precision	0.714286	1.0	0.911765	0.998390
Specificity_TNR	0.934509	1.0	0.987406	0.999640
model_f1	0.711679	1.0	0.696629	0.981013
FPR	0.065491	0.0	0.012594	0.000360